As in other organs, 2 areas of inquiry into CNS dysfunction (particularly that due to inherited metabolic or neurogenetic disease) have recently converged: neural progenitor cell biology & CNS via transplantation. Through retrovirus-mediated gene transfer, we previously generated immortalized, clonal, multipotent murine neural progenitor lines. Examination of their differentiation potential in vitro suggested enormous plasticity at the level of the individual progenitor. Upon transplantation, these immortalized progenitors engrafted in a non- tumorigenic, cytoarchitecturally & functionally appropriate manner, recapitulating their multipotency in vivo & expressing retrovirally- transduced genes in a robust fashion within brain parenchyma for prolonged periods. Because the blood-brain barrier imposes restrictions to entry of enzyme supplied peripherally (either directly or through genetically-engineered somatic cells) & because bone marrow transplantation entails irradiation which is inimical to developing CNS, peripheral treatment of CNS manifestations of genetic disease has been disappointing. Delivery of gene products directLy to CNS would circumvent such problems. (Grafting primary fetal tissue for such purposes poses numerous biologic & ethical concerns). Our data suggest the feasibility of transplanting immortalized neural progenitors constituitively expressing missing gene products, or genetically engineered to do so, as a strategy for sustained therapy of CNS manifestations of neurovisceral disease, &/or to effect repair as integral members of CNS cytoarchitecture. This proposal attempts to establish a paradigm of neural precursor transplantation as a therapy for CNS insult through 4 aims: (1) Confirm preliminary findings that a given clonal progenitor line can engraft (with technical ease) & participate normally in the development of multiple structures along the neuraxis & at multiple stages spanning from embryo to adult, that the progenitors can differentiate into multiple cell types in response to prevailing spatial & temporal cues, & express their retrovirally-transduced reporter gene. (This would broaden enormously the applications of such an immortalized precursor.) (2) Insure consistently efficient engraftment & safety of recipient animals by identifying the variables which direct engraftment, & determining the properties & fate of cells in situ that do engraft & express transgenes. Grafted cells can be traced & characterized in vivo; they can be retrieved & re-examined in vitro; their efficiency of engraftment vs gene expression can be quantified; their impact on host brain can be assessed. (Such characterization is crucial prior to consideration of similar strategies in humans.) (3) Transfer an exogenous gene or factor, via transplantation of an expressing precursor, into the CNS of a prototypical mouse model of neurovisceral disease in which a defect in that gene has been defined. (4) Transplant progenitors into a lesioned or mutant mouse model of a cell type-specific neurodegeneration to determine if they will integrate into the CNS & assume the phenotype of the deficient cell-type. Pilot data attest to the feasibility of each aim.